The present invention is generally in the field of methods for manufacture of recombinant proteins, and especially in the field of refolding of recombinant proteins expressed in the inclusion bodies of procaryotic expression systems such as E. coli. 
Expression of recombinant proteins with natural biological activity and structure, referred to as xe2x80x9cproteomicsxe2x80x9d, becomes increasingly important with the completion of genomic sequencing for several organisms and the near completion of human genome sequencing. One aspect of proteomics is to express large amounts of protein for structural and functional studies, as well as for commercial applications. The least expensive and most efficient way to express recombinant proteins is to express the proteins in E. coli. Proteins are expressed either intracellularly or secreted into the periplasmic spaces. In the former case, the proteins are often deposited in inclusion bodies, especially if the protein has disulfide bonds.
However, one of the problems in expressing mammalian proteins in E. coli is that most of the expressed proteins form insoluble inclusion bodies. While this problem can be circumvented by using various mammalian or insect expression systems, growing E. coli is faster and less expensive compared to mammalian and insect cultures. Moreover, some proteins are toxic to the host when expressed in their native forms, thus expression as insoluble inclusion bodies is the only way to obtain large quantities of recombinant proteins. Importantly, high levels of expression can be achieved for most proteins. 400 to 600 mg of inclusion bodies per liter of bacterial culture can routinely be achieved, with up to 9,700 mg/L having been reported using this method (Jeong KL; Lee SY, 1999. Appl. Environ. Microbiol. 65:3027-32). Inclusion bodies can be easily purified to greater than 90% with a simple freeze/thaw and detergent washing procedure.
Inclusion bodies appear as dense cytoplasmic granules when the cells are observed under a light microscope. Typically, the cells will be lysed by mechanical disruption of the cells, followed by centrifugation for 30 min at 4700 g. Inclusion bodies will sediment at low g forces and can be separated from many other intracellular proteins. Further purification can be done by washing the pellet with the buffer used during the cell disruption, or by centrifuging the resuspended pellet in 40-50% glycerol.
Many extracellular proteins of eukaryotes contain disulfide bonds. Proteins having multiple disulfide bonds may form non-native disulfide bonds during folding from the reduced species. Further folding is then blocked unless the incorrect disulfide bond is cleaved by reduction with an external thiol or by attack from a protein thiol. Eukaryotic organisms that secrete disulfide containing proteins also machinery for ensuring proper disulfide bond formation. A distinct disadvantage of expression of recombinant proteins in prokaryotes as inclusion bodies is that the proteins are not obtained in their native state, and typically are not functionally active. A variety of methods have been used to re-solubilize the proteins and refold them to reform active protein. Dissolution of the pelleted recombinant protein usually requires the use of denaturants such as 7 M guanidine hydrochloride or 8 M urea. The amount of aggregation may continue to increase with time if the protein is allowed to remain in the denaturant (Kelley and Winkler, xe2x80x9cFolding of Eukaryotic Proteins Produced in Escherichia colixe2x80x9d Genetic Engineering 12, 1-19 at p. 6 (1990)). Removal of the denaturant from the solubilized inclusion bodies by dialysis or desalting columns will cause the protein to precipitate under conditions where the native protein needs to be refolded. A misfolded protein solution can also have a very low specific activity in biological assays.
Although there are many reports of expression and refolding of various proteins in E. coli as inclusion bodies, one of the misconceptions in protein refolding is that a unique refolding method has to be developed for each individual protein (see Kelley and Winkler at p. 6). Another misconception is that most of the mammalian proteins cannot be refolded from inclusion bodies. (for review, see: Rudolph R., Lilie H., 1996. FASEB J 10:49-56; Lilie H, Schwarz E, Rudolph R. 1998. Curr Opin Biotechnol 9:497-501). Because published works are mostly xe2x80x9csuccessxe2x80x9d stories in refolding inclusion bodies from E. coli, it is impossible to get a general idea about what percentage of mammalian proteins can be purified using this procedure.
There are probably more refolding methods than refolded proteins reported in the literature (for review, see: Rudolph R., Lilie, H. 1996, FASEB J 10:49-56; Lilie, H., Schwarz, E., Rudolph, R. 1998, Curr. Opin. Biotechnol. 9:497-501). Different chaperones, detergents, and chaotrophs have been used to help refolding. In addition, pH, ionic strength, temperature, buffer formulation, and reducing/oxidation reagents can all effect refolding. It would be prohibitive to test all these conditions for refolding large amounts of proteins, as required for studies in proteomics or structural genomics.
A single simplified procedure to refold most of the proteins that are expressed in recombinant systems, especially those which form inclusion bodies in systems such as E. coli, is therefore needed.
It is therefore an object of the present invention to provide a xe2x80x9cuniversalxe2x80x9d method for refolding of proteins, especially recombinant proteins, especially recombinant proteins present in inclusion bodies in bacterial hosts.
A universal folding method that has been demonstrated to be effective in refolding a variety of very different proteins expressed in bacteria as inclusion bodies has been developed. Representative proteins that can be dissolved and refolded in biologically active form, with the native structure, are shown in Table I. The method has two key steps to unfold and then refold the proteins expressed in the inclusion bodies. The first step is to raise the pH of the protein solution in the presence of denaturing agents to pH greater than 9, preferably 10. The protein solution may be maintained at the elevated pH for a period of up to about 24 hours, or the pH immediately decreased slowly, in increments of about 0.2 pH units/24 hours, until the solution reaches a pH of about 8.0, or both steps used. In the preferred embodiment, purified inclusion bodies are dissolved in 8 M urea, 0.1 M Tris, 1 mM glycine, 1 mM EDTA, 10 mM beta-mercaptoethanol, 10 mM dithiothreitol (DTT), 1 mM redued glutathion (GSH), 0.1 mM oxidized glutathion (GSSG), pH 10. The absorbance at 280 nm (OD280) of the protein solution is 5.0. This solution is rapidly diluted into 20 volumes of 20 mM Tris base. The resulting solutin is adjusted to pH 9.0 with 1 M HCl and is kept at 4xc2x0 C. for 24 hr. The pH is adjusted to pH 8.8 and the solution is kept at 4xc2x0 C. for another 24 hrs. This process is repeated until the pH is adjusted to 8.0. After 24 hr at pH 8.0, the refolded proteins can be concentrated by ultrafiltration and applied to a gel filtration column for purification.
Recombinant proteins are typically expressed in a suitable host, for example, a procaryotic expression system such as E. coil or other type of bacteria, using a standard expression vector like a plasmid, bacteriophage or even naked DNA, and the protein expressed from the plasmid or DNA integrated the host chromosome. Suitable bacterial strains are commercially available or can be obtained from the American Type Culture Collection (ATCC), 10801 University Boulevard, Manassas, Va. 20110-2209.
Suitable vectors can be obtained from any number of sources, including the ATCC. These need a promoter to insure that the DNA is expressed in the host, and may include other regulatory sequences. The vector may also include means for detection, such as an antibiotic resistance marker, green fluorescent protein tag, or antigen tag to facilitate in purification of the recombinant protein.
Once the DNA encoding the protein to be purified is introduced into the host, the host is cultured under appropriate conditions until sufficient amounts of recombinant protein are obtained.
Once the protein has been expressed in the maximum amount, it must be separated and purified from the bacterial host. The protein is isolated generally by lysing the cells, for example, by suspending in detergent, adding lysozyme, and then freezing (for example, by suspending cells in 20 ml of TN/1% Triton(trademark) X-100, adding 10 mg lysozyme and freezing at xe2x88x9220xc2x0 C. overnight), thawing and adding DNAase to degrade all of the bacterial DNA, then washing the resulting precipitate in a buffered solution. The precipitate is then dissolved in an appropriate solution as discussed below, for refolding.
The isolated protein is then refolded. There are several critical aspects of an universal refolding method.
(i) High pH refolding. Most published procedures refold proteins using reducing chaotrophs (such as 8 M urea) at a physiological pH, usually pH 7.4 to 8.0. This usually produces large quantities of precipitation or aggregation, making refolding either impossible or with a very low yield. It has been found that some proteins cannot be refolded at physiological pH, but can be refolded when initial refolding pH is high (at least pH 9.0, although higher pH, such as pH 10, may be desirable). This strategy was initially inspired by the fact that pepsinogen can be reversibly denatured/renatured between pH 8.0 and 9.0. It is postulated that at high pH (such as pH 9.0), proteins can obtain some secondary structures, allowing it to be refolded more efficiently when acidity of the refolding solution is lowered to the biological pH. Later it was found that even for the proteins that could be refolded at a physiological pH, high pH refolding resulted in a better yield. In addition, high pH refolding is an excellent method for preventing initial large scale precipitation.
(ii) Non-denaturing chaotroph concentration. It has been shown in several labs that non-denaturing concentrations of chaotrophic reagents, such as 0.5 to 1.0 M of urea, guanidine hydrochloride, and L-arginine, can be used to assist refolding and stabilize refolded proteins (Rudolph et al. 1996, FASEB J 10:49-56). The preferred concentration of chaotrophic reagents is 0.4 M, although the concentration may range from 0 to 4 M. A moderate concentration of urea in the refolding/purification procedures has not been found to have a denaturing effect on proteins.
(iii) Reducing/oxidation reagents. Inclusion bodies for mammalian proteins containing disfulfide bonds need to be dissolved in the presence of reducing reagents. Representative reducing agents include beta-mercaptoethanol, in a range of from 0.1 mM to 100 mM, preferably 10 mM; DTT, in a range of from 0.1 mM to 10 mM, preferably 10 mM; reduced glutathion (GSH), in a range from 0.1 mM to 10 mM, preferably 1 mM; and oxidized glutathion (GSSG), in a range from 0.1 mM to 10 mM, preferably 1 mM. Beta-mercaptoethanol is a preferred reducing reagent. In addition, dithiothreitol and/or reduced/oxidized glutathion (GSH, GSSG) can also be included to facilitate xe2x80x9coxido-shufflingxe2x80x9d of wrongly folded, intermediate disulfide bonds.
(iv) pH control It is important that the protein solution remain at an elevated condition long enough to refold the protein. This is preferably achieved by decreasing the pH slowly, in 0.2 pH unit increments per 24 hours. In this method, the protein solution is adjusted to a high pH, preferably at least 9.0 or higher, to 10 or less preferably 11. The protein is preferably maintained at each pH for at least 24 hours, although comparable effects can be achieved with shorter periods of time, for example, for a period of three, six, nine, twelve, eighteen or twenty hours, most preferably at least twelve hours. The pH can be adjusted by addition of an acid or by dialysis or dilution into a lower pH. Addition of the acid is preferred.
The four conditions discussed above are considered the most essential aspects of the basic protocol for the xe2x80x9cuniversalxe2x80x9d refolding procedure.
Lin, X, Wong, R. N. S., and Tang, J. (1989) xe2x80x9cSynthesis, purification, and active site mutagenesis of recombinant porcine pepsinogenxe2x80x9d. J. Biol. Chem. 264:4482-4489.
Lin, X. L., Lin, Y. -Z., Koelsch, G., Gustchina, A., Wlodawer, A., and Tang, J. (1992) xe2x80x9cEnzymic activities of two-chain pepsinogen, two-chain pepsin, and the amino-terminal lobe of pepsinogenxe2x80x9d. J. Biol. Chem. 267:17257-17263.
Lin, X., Loy, J. A., Sussman, F., and Tang, J. (1993) xe2x80x9cConformational instability of the N- and C-terminal lobes of porcine pepsin in neutral and alkaline solutionsxe2x80x9d. Prol. Sci. 2:1383-1390.
Chen, Z., Koelsch, G., Han, H. -P., Wang, X. -J., Lin, X., Hartsuck, J. A., and Tang, J. (1991) xe2x80x9cRecombinant rhizopuspepsinogenxe2x80x9d, J. Biol. Chem. 266:11718-11725.
Lin, Y. -Z., Fusek, M., Lin, X. L., Hartsuck, J. A., Kezdy, F. J., and Tang, J. (1992) xe2x80x9cpH Dependence of kineticparameters of pepsin, rhizopuspepsin, and their active-site hydrogen bond mutantsxe2x80x9d. J. Biol. Chem. 267:18413-18418.
Lin, X. L., Liu, M. T., and Tang, J. (1992) xe2x80x9cHeterologous expression of thermopsin, a heat stable acid proteinasexe2x80x9d. Enzyme Microb. Technol. 14:696-701.
Lin, Y. -Z., Lin, X., Hong, L., Foundling, S., Heinrikson, R. L., Thaisrivongs, S., Leelamanit, W., Raterman, D., Shah, M., Dunn, B. M., and Tang, J. (1995) xe2x80x9cEffect of point mutations on the kinetics and the inhibition of human immunodeficiency virus type 1 protease: relationship to drug resistancexe2x80x9d. Biochemistry 34:1143-1152.
Ermolieff, J., Lin, X., and Tang, J. (1997) Kinetic properties of saquinavir-resistant mutants of human immunodeficiency virus type 1 protease and their implications in drug resistance in vivo. Biochemistry 36:12364-12370.
Lin, X., Tang, J., Koelsch, G., Monod, M., and Foundling, S. (1993) xe2x80x9cRecombinant canditropsin, an extracellular aspartic protease from yeast Candida tropicalisxe2x80x9d. J. Biol. Chem. 268:20143-20147.
Koelsch, G., Tang, J., Monod, M., Foundling, S. I., Lin, X. (1998) xe2x80x9cPrimary substrate specificities of secreted aspartic proteases of Candida albicansxe2x80x9d. Adv. Exp. Med. Biol. 436:335-338.
Wang, X., Lin, X., Lowy, J. A., Tang, J., Zhang, X. C. (1998) xe2x80x9cCrystal structure of the catalytic domain of human plasmin complexed with streptokinasexe2x80x9d, Science. 281:1662-1665.
Wang, X., Terzyan, S., Tang, J., Loy, J., Lin, X., and Zhang, X. (2000) xe2x80x9cHuman plasminogen catalytic domain undergoes a novel conformational change upon activationxe2x80x9d J. Mol. Biol. (in press).
Faro, C., Ramalho-Santos, M., Vieira, M., Mendes, A., Simoes, I., Andrade, R., Verissimo, P., Lin, X., Tang, J., Pires, E. (1999) xe2x80x9cCloning and Characterization of cDNA Encoding Cardosin A, an RGD-containing Plant Aspartic Proteinase,xe2x80x9d J. Biol. Chem. 274(40):28724-28729.
Lin, X., Koelsch, G., Wu, S., Downs, D., Dashti, A., and Tang, J. (2000) xe2x80x9cHuman aspartic protease memapsin 2 cleaves the xcex2-secretase site of xcex2-amyloid precursor protein. Proc. Natl. Aca. Sci. 97(4):1456-1460.
Ghosh, A. K., Shin, D., Downs, D., Koelsch, G., Lin, X., Ermolieff, J., Tang, J. (2000) xe2x80x9cDesign of potent inhibitors form human brain memapsin 2 (xcex2secretase)xe2x80x9d J. Amer. Chem. Soc., 122:3522-3523.
The present invention will be further understood by reference to the following non-limiting examples.